linking ASEC and Job Tenure Supplement at person level

Is it possible to link ASEC data with Job Tenure Supplement data at the individual level? I’ve read about the various ASEC issues with CPSIDP and know you’ve now got the MARBASECIDP variable available to enable linking with March Basic, but can’t figure out whether it’s possible to somehow bridge those to link ASEC with Job Tenure for individuals? If so, how would you recommend going about it?

And if not, would a combo of HRHHID/HRHHID2/LINENO/SEX be a reasonable alternative? Any major downsides to that?

Thanks so much!

In theory it should be possible to link individuals in the ASEC samples to households in the Job Tenure Supplement samples. You first need to link the ASEC with its corresponding March basic monthly sample using MARBASECIDP. You could then link this with the Job Tenure Supplement (which is typically enumerated in January) using CPSIDP. When performing this second linking, please take note of some quirks of the CPS sampling scheme as described on the CPSIDP description tab. Namely the 4-8-4 rotational sampling pattern and to verify linkages with AGE, SEX, and RACE.

I hope this is helpful.

Hi Jeff,
Can you explain why one needs to first link ASEC to March BMS?

The quote below is from this page: IPUMS CPS

“However, it is not necessary to first link the ASEC to the March BMS before linking the ASEC information to other months; CPDIDP will make all possible links on its own.”

I’m trying to link the job tenure supplement to the ASEC and I don’t see why I need MARBASECIDP rather than just using CPSIDP.

thanks!!

The page you linked to is correct. It is no longer necessary to go through the intermediate step of using MARBASEC since CPSIDP alone allows you to link the ASEC to other months of the CPS (such as the biennial January job tenure supplement). As Jeff points out, CPSIDP links aren’t perfect and it’s important to verify them using variables such as AGE, SEX, and RACE. I would recommend taking a look at the linking exercises and sample Stata code on this page for further reference. An additional feature to be aware of is that individuals who are part of the ASEC oversample are unable to be linked with the basic survey months and will have a CPSIDP value of 0.